Trastuzumab deruxtecan (T-DXd) vs ramucirumab (RAM) + paclitaxel (PTX) in second-line treatment of patients (pts) with human epidermal growth factor receptor 2-positive (HER2+) unresectable/metastatic gastric cancer (GC) or gastroesophageal junction adenocarcinoma (GEJA): Primary analysis of the randomized, phase 3 DESTINY-Gastric04 study.

LBA4002 Background: T-DXd 6.4 mg/kg is approved for pts with metastatic HER2+ GC/GEJA who received a prior trastuzumab-based regimen based on prior phase 2 studies. This is the primary efficacy analysis from a planned interim analysis of DESTINY-Gastric04 (NCT04704934), a global, randomized, multicenter, open-label, phase 3 study evaluating the efficacy and safety of T-DXd vs RAM + PTX in pts with HER2+ unresectable/metastatic GC/GEJA in this second-line setting. Methods: After biopsy-confirmed HER2+ status (IHC 3+ or IHC 2+/ISH+), pts were randomized 1:1 to T-DXd 6.4 mg/kg or RAM + PTX. The primary endpoint was overall survival (OS). OS between the 2 arms was compared by a log-rank test stratified using randomization factors. Secondary endpoints by investigator assessment include progression-free survival (PFS), confirmed objective response rate (cORR), disease control rate (DCR), and safety. Results: At data cutoff (October 24, 2024), 494 pts were assigned (T-DXd, n = 246; RAM + PTX, n = 248). Based on 266 OS events observed (information fraction = 78.5%), efficacy superiority was achieved (2-sided P < 0.0228). Median (m) (95% CI) OS follow-up was 16.8 mo (14.0-20.0) for T-DXd and 14.4 mo (13.1-19.7) for RAM + PTX. mOS (95% CI) was 14.7 mo (12.1-16.6) for T-DXd vs 11.4 mo (9.9-15.5) for RAM + PTX (hazard ratio [HR], 0.70; P = 0.0044). Additional efficacy data are in the Table. Median (range) treatment duration was 5.4 mo (0.7-30.3) with T-DXd and 4.6 mo (0.9-34.9) with RAM + PTX. Treatment-emergent adverse events (TEAEs) were reported in 244/244 (100%) vs 228/233 pts (97.9%) with T-DXd vs RAM + PTX, respectively; 68.0% vs 73.8% were grade (G) ≥3. Serious TEAEs with T-DXd vs RAM + PTX occurred in 41.0% vs 43.3% of pts; TEAEs associated with drug discontinuation occurred in 14.3% vs 17.2% of pts. Independently adjudicated drug-related interstitial lung disease/pneumonitis occurred in 34 pts (13.9%) with T-DXd (1 G3, 0 G4/5) vs 3 pts (1.3%) with RAM + PTX (2 G3, 1 G5). Conclusions: T-DXd showed statistically significant and clinically meaningful improvement in OS over RAM + PTX in pts with HER2+ unresectable/metastatic GC/GEJA, reinforcing its use as a second-line standard of care. The safety profile of T-DXd 6.4 mg/kg was consistent with the known safety profile of T-DXd in GC/GEJA, with no new safety signals. Clinical trial information: NCT04704934 . Efficacy T-DXdn = 246 RAM + PTXn = 248 HR (95% CI) P value mOS (95% CI), mo 14.7 (12.1-16.6) 11.4 (9.9-15.5) 0.70 (0.55-0.90) P = 0.0044 mPFS (95% CI), mo 6.7 (5.6-7.1) 5.6 (4.9-5.8) 0.74 (0.59-0.92) P = 0.0074 cORR (95% CI), % 44.3 (37.8-50.9) 29.1 (23.4-35.3) P = 0.0006 DCR (95% CI), % 91.9 (87.7-95.1) 75.9 (70.0-81.2)